The Radar

What I'm Watching

This is what I'm actually paying attention to. Not what's trending. Not what got applause on Twitter. The real signals. Products shipping interesting decisions. Interfaces that make me stop and reverse-engineer why they work. Design moves that most people scroll past but quietly change how something feels.

I update this constantly because taste is a muscle. This is how I train mine. Some of it is praise. Some of it is me pointing at something and explaining why it's broken. Both are useful. If you want to see what a design operator actually watches when nobody's performing, you're looking at it.

↗ The Death of the Dashboard

Every SaaS product built in the last decade starts with a dashboard. Charts, numbers, widgets, a greeting that says "Welcome back." Most users glance at it and immediately navigate somewhere else. The dashboard became a parking lot, not a launchpad.

The smartest AI-native products are killing it. Instead of showing you a snapshot of data, they're opening with a single question: "What do you want to do?" or surfacing the one thing that changed since you were last here.

This is a fundamental shift in product design. The old model was: here's everything, go find what matters. The new model is: we already know what matters, here it is. The products that get this right will make dashboards feel like loading screens. A relic from when software couldn't think.

Watch for this pattern. It's going to eat SaaS.

↗ Wallet Connect Is Still the Biggest Lie in Web3 UX

The entire Web3 industry has accepted wallet connect as the front door. It shouldn't be. It's the equivalent of asking for a credit card before showing the menu. And the numbers prove it. Products that require wallet connection as step one see 30-50% drop-off at that screen alone.

The few teams getting this right are delaying the connect step. They let users explore. See returns. Interact with sample data. Build context. Then ask for the wallet. Not because they're being clever. Because they understand that trust is sequential. You don't ask for commitment before you've demonstrated value.

The wallet connect screen isn't a UX problem. It's a trust sequencing problem. Every Web3 product that solves this will outperform competitors who don't, even if the underlying product is identical.

↗AI Products That Hide the AI Are Winning

There's a split forming in AI product design. On one side, products that lead with "AI-powered" in the headline, put sparkle icons on every button, and build their brand around the technology. On the other side, products where you don't even notice AI is involved. The experience just feels faster, smarter, and strangely anticipatory.

The second group is winning. Not because AI branding is bad. Because most users don't care about the technology. They care about the outcome. "AI-powered" is a feature description. "This saves you three hours" is a value proposition.

The best AI interfaces right now feel invisible. The AI makes decisions behind the scenes: pre-filling forms, predicting next actions, surfacing relevant context before you search for it. No badges. No sparkle animations. Just an experience that quietly removes friction the user didn't even know was there.

Products that make AI invisible will own the next decade. Products that make AI the headline will get replaced by them.

↗ The Return of the One-Page Website

Multi-page marketing sites are losing to single-page, high-conviction landing pages. Not because attention spans are shorter. Because decision-making is faster when there's less to process.

The pattern: one headline, one subline, one CTA above the fold. Social proof immediately below. One section explaining how it works. One section on who it's for. Pricing. Done. No hamburger menus. No blog link. No "About Us" page that takes you out of the conversion flow.

This isn't lazy. It's disciplined. Every page you add to a marketing site is another exit point. Every navigation link is an invitation to wander instead of decide. The founders who understand this are shipping landing pages that feel more like billboards than websites. And they're converting at rates that make multi-page sites look bloated.

If your marketing site has more than five pages and you're pre-Series A, you probably have three pages too many.

↗ Design Systems Are Becoming the Product

Something interesting is happening with design systems. They're moving from internal tooling to external product. Companies are open-sourcing their systems not as charity but as distribution strategy. When your design system becomes the default for an ecosystem, you own the visual language of everything built on top of you.

This matters more in Web3 and AI than anywhere else. Both ecosystems are moving fast, and most teams don't have the bandwidth to build their own component infrastructure from scratch. The protocols and platforms that ship opinionated, high-quality design systems will see their ecosystem products look and feel cohesive by default. That's not just good design. That's platform strategy.

If you're building a protocol or platform and you haven't thought about your design system as a distribution tool, you're leaving ecosystem cohesion and developer adoption on the table.

↗ Brutalist Web Design Is Not a Trend. It's a Reaction.

Brutalist design keeps being called a trend. It's not. It's a reaction to a decade of everything looking the same. When every SaaS landing page uses the same gradient, the same rounded corners, the same illustration style, the same Figma template energy, anything that breaks that pattern gets attention by default.

The real signal here isn't aesthetic. It's strategic. The products and brands reaching for brutalism are making a positioning choice. They're saying: we're not like everyone else, and we want you to feel that the second you land. It's differentiation through visual tension.

The mistake is copying the aesthetic without understanding the strategy. Brutalism works when it's intentional. When it's just another template, it becomes the thing it was reacting against.

↗ Micro-Interactions Are the New Brand Language

Logos used to carry most of the brand weight in digital products. Colors and typography carried the rest. Now, the brands that feel most alive are the ones investing in how things move.

A button that bounces slightly on press. A loading state that feels playful instead of empty. A transition between screens that has just enough personality to feel intentional. These micro-interactions are becoming the primary way users experience brand in digital products. Not because motion is new. Because everything else has converged. When two products have the same layout, the same type scale, and the same color palette, the one that moves better feels better.

This is especially true in mobile. On a six-inch screen with minimal visual real estate, motion is the differentiator. The brands investing here are building emotional connections that static design can't replicate.

↗ The Creator Tool Explosion and Why Most Will Fail

You can change every color, every image, and every layout on a website and it will feel like the same product. Change the typeface and it becomes a different brand. Typography is the single highest-leverage design decision most teams spend the least time on.

The gap is widening. The best brands are commissioning custom type or using carefully selected typefaces as core brand elements. The rest are defaulting to Inter or whatever system font ships with their template.

This matters because typography is subconscious. Users don't think "nice typeface." They think "this feels premium" or "this feels trustworthy" or "this feels like every other startup I've seen this week." The typeface is doing that work silently. Most teams never realize it.

If your product feels generic and you've changed everything else, change the type. It's probably the thing you haven't touched.

↗ Typography Is Still the Most Undervalued Design Decision

Everyone is building AI-powered creative tools right now. Image generators. Writing assistants. Video editors. Music producers. Design copilots. The market is flooded.

Most will fail. Not because the AI isn't good enough. Because the products aren't opinionated enough. When you build a tool that does "everything," you've built a tool for nobody.

The creative tools that will survive are the ones that make a specific creative decision easier for a specific type of creator. Not "generate any image." Instead: "generate product photography that looks like.

↗ Dark Patterns Are Getting Smarter and That's a Design Ethics Problem

Dark patterns used to be obvious. The hidden unsubscribe link. The pre-checked box. The confusing cancellation flow. Now they're evolving into something harder to spot. Persuasive defaults buried inside AI recommendations. Personalized urgency that feels real but isn't. Interfaces that technically give you a choice but architecturally make one option invisible.

This matters because as AI gets better at predicting behavior, the line between personalization and manipulation gets thinner. A product that shows you "recommended for you" is helpful. A product that shows you "recommended for you" because it maximizes their revenue, not your satisfaction, is a dark pattern wearing a personalization costume.

The design industry needs to have a harder conversation about this. Not "dark patterns are bad." Everyone agrees with that. The real question is: where does smart personalization end and manipulation begin? And who decides?